Scikit-Learn & TensorFlow练习题问答第8章

Why do we reduce a dataset's dimensionality? What are the primary motives and limitations of such a process?

A1:

Motivations:

• 为了提升后续训练算法的速度
• 通过可视化数据来获取关键特征的见解
• 为了节省存储空间类似于压缩

Drawbacks:

• Particular information may be lost, which could potentially diminish the performance of subsequent training algorithms.
  • Serving its purpose, this method requires significant computational resources.
  • Incurring some complexity, this approach increases complexity in your Machine Learning pipelines.
  • The transformed features are frequently challenging to interpret and reconstruct.

Q2. Wh